Vultr Vultr Dedicated Cloud vs Vultr Performance VPS Benchmarks

Vultr just announced their new Dedicated Cloud series offerings. So I am comparing their 2cpu 4GB Performance VPS versus their 2cpu 8GB Dedicated Cloud series and tabulating the inclusion of previous Vultr 2GB VPS tests for comparison. So I thought I would revisit comparison benchmarks with my centminmodbench.sh script and Centmin Mod .08 beta 03 LEMP web stack environment.

Previous DigitalOcean vs Vultr vs Linode benchmark comparison was with just their normal Performance Series VPS instances. VPS Performance instances seem to be now on 2.4Ghz cpus if you see the bogomips rating of 4800 still shows on VPS series tested below and matches with my previous test of 2.4Ghz Vultr VPS.

Servers:

---

All servers are installed with my own Centmin Mod .08 beta 03 LEMP web stack environment consisting of:

• CentOS 6.6 64 bit
• Nginx 1.9.1 with stream, SPDY/3.1 and ngx_pagespeed module support + vhost traffic stats module + Nginx Lua + LuaJIT support
• PHP 5.4.41 (PHP-FPM) with APC Cache 3.1.13
• MariaDB 10.0.19 MySQL
• Memcached 1.4.24 with memcache/memcached/libmemcached support
• CSF Firewall

For Vultr:

• Vultr KVM VPS Performance Series- 2.40Ghz New Jersey: 4GB, 2 CPUS, 90GB SSD 4TB (1Gbps network in/out) US$32-40/month
• Vultr Dedicated Cloud Series - 3.60Ghz New Jersey: 8GB, 2 CPUS, 120GB dedicated SSD disk 10TB (1Gbps network in/out) US$60/month
• Vultr KVM VPS - 2.40Ghz New Jersey: 2GB, 2 CPUS, 45GB SSD 3TB (1Gbps network in/out) US$20/month - from May 6th, 2015 tests
• Vultr KVM VPS - 3.60Ghz Los Angeles: 2GB, 2 CPUS, 45GB SSD 3TB (1Gbps network in/out) US$20/month - from May 6th, 2015 tests

Testing Methodology:

I basically created each respective server using CentOS 6.6 64bit OS and ran YUM update to bring them all inline with CentOS 6.6 64bit OS. Then ran the new installnbench.sh file at centminmod/centminmodbench · GitHub which I wrote specifically for this testing. It basically does 2 tasks when run installnbench.sh via curl command line on a fresh CentOS 6.x or 7.x OS server:

1. install latest Centmin Mod .08 beta03 LEMP stack ~ 15-30 mins time then (note .08 beta03 unlike previous benchmarks has now included some of the source downloads as part of the installer so you can not directly compare Centmin Mod .08 beta03 install times with previous and earlier .08 betas)
2. installs and runs centminmodbench.sh (UnixBench enabled) ~ 30-60 mins time

Code:

curl -sL https://github.com/centminmod/centminmodbench/raw/master/installnbench.sh | bash

Server Intel CPU Clock Speed:

The Vultr Dedicated Cloud server is using 3.60Ghz and the Vultr Performance VPS server is using 2.40Ghz instances apparently, so performance closely reflects that of my previous comparison of Vultr 3.60Ghz vs 2.40Ghz.

Vultr Dedicated Cloud Series Setup

Vultr 2cpu 4GB Performance VPS (right) versus their 2cpu 8GB Dedicated Cloud series (left)

 

Centmin Mod .08 beta 03 LEMP Install Times:

Centmin Mod .08 beta03 LEMP web stack install times are highly variable as part of the time is downloading source tarballs or YUM package downloads and installs so depends on the speed of the respective YUM mirrors the system picks up to download from as well as location of the VPS server geographically to the downloads. Downloads account for between 15-33% of the total install time. The rest is source compilation for Nginx, PHP-FPM, PHP extensions, Memcached server, CSF Firewall and other installed items via the Centmin Mod auto installer. Ultimately, comes down to cpu processor, disk performance and cpu clock speed for source compilation parts.

For installnbench.sh users, you'll find the install time on the very last line in the /root/centminlogs/centminmod_1.2.3-eva2000.08_***-***_install.log install log.

Centmin Mod LEMP .08 beta 03 LEMP stack install has 2 parts, the YUM related download parts and the total install time which includes YUM, tarball download and source compilation. Centminmodbench.sh time is the total time it took to run the centminmodbench.sh script itself.

.08 beta03 changes

Note .08 beta03 unlike previous benchmarks has now included some of the source downloads as part of the installer so you can not directly compare Centmin Mod .08 beta03 install times with previous and earlier .08 betas. There's also additional metrics recorded during initial install for source download time, nginx and php-fpm first time source compilation installs.

Basically, you can see the pattern which will show throughout the below benchmarks. The new Vultr Dedicated Cloud instances are just the old Vultr Performance Series 3.60Ghz clocked VPS moving to a new line of product with a much higher price - almost 2-3x times the price. Seems intention for Vultr is for the Performance VPS to have the lower performing and cpu clocked 2.40Ghz now ?

 

centminmodbench.sh Results:

Ioping Disk and Ioping RamDisk memory test for random I/O performance.

Disk



Vultr Dedicated Cloud has less contention or noisy neighbours if you look at latency times compared to VPS servers. However, being 120GB SSD disks and the nature of NAND at that capacity it's expected that random disk I/O will be slower as you can see from above. Also it seems the new 2.40Ghz clocked Performance VPS series performs even slower than the 2.40Ghz clocked VPS back in May 6, 2015.

RamDisk - testing system memory bandwidth. Definitely some contention on Vultr 2CPU 4GB Performance VPS if you look at ioping seek max latency times. Again, the same pattern emerges, the new Vultr Dedicated Cloud seems to be basically the old 3.60Ghz clocked VPSes at a higher asking price.

 

DD Disk Benchmarks

Disk



Definitely much lower disk performance for current tests versus tests done back on May 6, 2015. The Vultr Dedicated Cloud using 120GB SSD definitely limited.

Ramdisk



Again you see the pattern that old 3.60Ghz VPS is basically the new Vultr Dedicated Cloud series with similar but higher memory bandwidth results.

 

Disk FIO benchmarks

Disk only test as not supported for ramdisk tmpfs.



Random reads for Vultr Dedicated Cloud and old 3.60Ghz VPS are about the same. Random writes are lower on both the current Vultr Dedicated Cloud and Vultr Performance series compared to the older VPS tested back in May 6, 2015.

 

Bandwidth Tests

Tested both single thread wget and multi-threaded parallel axel bandwidth tests. Bear in mind geographical location of VPSes in relation to bandwidth numbers and that Vultr 2GB 3.60Ghz is located in Los Angeles compared to the other 3 in New York/New Jersey.

wget bandwidth



axel bandwidth

 

Ping Tests

Ping tests are all geographically relative to the VPS server location, so West coast USA would fair better with same location ping test sites and vice versa for East coast USA based Vultr NY VPS.

 

VPS OpenSSL 1.01e system benchmarks

Very clear that if you run a https SSL based site or are doing any type of OpenSSL encryption or decryption, that you do not want to be using Vultr 2.40Ghz VPS instances or even the new Vultr Performance VPS. Vultr Dedicated Cloud definitely with dedicated resources faired best with almost twice the performance where it counts for SSL related values like rsa 2048/4096 bit and 256bit ecdsa.

VPS Nginx static OpenSSL 1.02a benchmarks

If you plan on using more performant based ECC 256 bit based SSL certificates using ECDSA then Vultr 3.60Ghz is twice as fast as the rest and seems Vultr Dedicated Cloud fits that bill now. Centmin Mod .08 beta03 and higher might be switching from Nginx compiled with OpenSSL 1.02a to Nginx + LibreSSL later on. So curious about those static SSL numbers later on.

VPS Entropy Availability

Entropy availability pretty close seeing as Centmin Mod LEMP stack install by default tweaks and improves this thanks to using haveged.

 

Vultr MySQL Benchmarks

mysqlslap tool is included with all MySQL installs including Oracle MySQL, MariaDB MySQL and Percona MySQL. The custom mysqlslap parameters I used are just arbitrarily in nature and only test MyISAM performamce across 64 clients.

Actual test command used is:

MariaDB 10.0.19 MySQL is the default installed MySQL version in Centmin Mod .08 beta03 LEMP auto installer.

mysqlslap results

Vultr 3.60Ghz instance comes out on top for new Vultr Dedicated Cloud with dedicated resources followed by the older 3.6Ghz VPS. Processor performance and clock speed plays a factor. MySQL slap though doesn't reflect real world usage so bear that in mind.

 

PHP benchmark

• PHP benchmarks test the Centmin Mod .08 beta03 installed PHP 5.4.41 (php-fpm) binary (which defaults include enabling APC Cache 3.1.13 with igbinary support) and tests against the PHP 5.4.41 source included Zend/bench.php and Zend/micro_bench.php files.
• 3 runs are made and average taken for time it takes to run the benchmarks.
• This test is single threaded in nature so doesn't scale with multiple cpu cores/threads and the underlying performance is based partly on cpu processor performance and clock speed. Which the final results confirm as Vultr 3.60Ghz is the clear winner with new Vultr Dedicated Cloud and older 3.6Ghz VPS managing 1st and 2nd place.

PHP configure options

 

New Vultr Dedicated Cloud and VPS UnixBench Benchmarks

Vultr Dedicated Cloud and VPS

Code:

uname -r
2.6.32-504.16.2.el6.x86_64

Vultr VPS kernels May 6, 2015 tested

Code:

uname -r
2.6.32-504.16.2.el6.x86_64

Vultr servers all used the same 2.6.x linux kernel, so you can see definitely compare UnixBench numbers across servers. To compare with non-2.6.x kernels, you can probably only rely on looking at the Dhrystone and Whetstone numbers to gauge cpu processor performance.

• Vultr Dedicated Cloud with dedicated resources and 3.60Ghz clock speed came out on top with older 3.60Ghz VPS coming second. The new 2CPU 4GB VPS was slower than that of 2CPU 2GB VPS tested back in May 6, 2015.

You can read the explanation at byte-unixbench - A Unix benchmark suite - Google Project Hosting which suggests subtests like pipe throughout has no real world bearing on performance so can be disregarded

 

Vultr Dedicated Cloud vs Vultr Performance VPS Summary

The reason why I did the above comparison benchmarks is because I constantly test my Centmin Mod LEMP menu based auto installer web stack. So it's good to know which VPS cloud hosting provider has the best performance for my needs which includes Centmin Mod install times (I may install and re-install 100s of times per month). As well as general performance that has some relative real world applications for what a normal LEMP web stack would be for - Nginx SPDY/3.1 SSL, PHP-FPM performance and MariaDB MySQL performance.

For Centmin Mod install times:

• Vultr 3.60Ghz is clearly fastest whether it be the new Vultr Dedicated Cloud or older 3.6Ghz Vultr Performance VPS. If you can grab an old 3.6Ghz grab one!

For Disk Performance:

• There's definitely less contention and noisy neighbours with the new Vultr Dedicated Cloud however the 120GB SSD disk capacities may limit write speed and random disk I/O. Would be much better if they bumped it up to 240/256GB SSD capacities.

For Bandwidth & Pings performance:

• Geographically relative

For OpenSSL performance:

• Nginx SPDY SSL and https performance is more important now given Google's announcement. So if you are planning to use Nginx SPDY/3.1 SSL based web server for your sites and forums, you would want to pick a VPS provider with the better Intel Xeon E5 v3 based processors. Vultr 3.60Ghz has the best AES-NI assisted aes128/aes256 performance especially for the new Vultr Dedicated Cloud.
• For SSL certificates related usage, the most commonly used key algorithms will be either rsa 2048 bit, rsa 4096 bit and 256 bit ecdsa. With that in mind, Vultr 3.60Ghz will have the edge on the new Vultr Dedicated Cloud compared to current 2.40Ghz clocked Vultr Performance VPS series.

For mysqlslap MySQL performance:

• Again dedicated resources and higher 3.60Ghz clock speed allow the new Vultr Dedicated Cloud to lead.

For PHP performance

• The PHP benchmarks were only single threaded so reflects more cpu processor performance and clock speed. Again Vultr 3.60Ghz on new new Vultr Dedicated Cloud and older 3.60Ghz clocked VPS lead the way.

For UnixBench performance:

• On the whole for cpu performance, the new Vultr Dedicated Cloud and older 3.60Ghz VPS are nearly twice as fast as the current 2.40Ghz clocked Vultr Performance VPS series.
• Below are the lscpu outputs for all 4 VPS servers.

new Vultr Dedicated Cloud

Code:

----------------------------------------------
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                2
On-line CPU(s) list:   0,1
Thread(s) per core:    1
Core(s) per socket:    1
Socket(s):             2
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 60
Stepping:              1
CPU MHz:               3600.012
BogoMIPS:              7200.02
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              4096K
NUMA node0 CPU(s):     0,1

Vultr 2CPU 4GB Performance VPS

Code:

----------------------------------------------
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                2
On-line CPU(s) list:   0,1
Thread(s) per core:    1
Core(s) per socket:    1
Socket(s):             2
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 60
Stepping:              1
CPU MHz:               2399.996
BogoMIPS:              4799.99
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              4096K
NUMA node0 CPU(s):     0,1

Vultr 3.60Ghz on May 6th, 2015

Code:

----------------------------------------------
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                2
On-line CPU(s) list:   0,1
Thread(s) per core:    1
Core(s) per socket:    1
Socket(s):             2
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 60
Stepping:              1
CPU MHz:               3600.014
BogoMIPS:              7200.02
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              4096K
NUMA node0 CPU(s):     0,1

Vultr 2.40Ghz on May 6th, 2015

Code:

----------------------------------------------
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                2
On-line CPU(s) list:   0,1
Thread(s) per core:    1
Core(s) per socket:    1
Socket(s):             2
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 60
Stepping:              1
CPU MHz:               2399.996
BogoMIPS:              4799.99
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              4096K
NUMA node0 CPU(s):     0,1

 

VPS Prices:

New plans and pricing matrix for Vultr compared with Linode and DigitalOcean current pricing:

Linode Pricing Plans

Web Host	CPU Cores	Memory	SSD Disk	Bandwidth	Network In	Network Out	Price /hr	Price /month
Linode	1	1GB	24GB	2TB	40Gbps	125Mbps	$.015/hr	($10/m)
Linode	2	2GB	48GB	3TB	40Gbps	250Mbps	$.03/hr	($20/m)
Linode	4	4GB	96GB	4TB	40Gbps	500Mbps	$.06/hr	($40/m)
Linode	6	8GB	192GB	8TB	40Gbps	1000Mbps	$.12/hr	($80/m)
Linode	8	16GB	384GB	16TB	40Gbps	2000Mbps	$.24/hr	($160/m)
Linode	12	32GB	768GB	20TB	40Gbps	4000Mbps	$.48/hr	($320/m)
Linode	16	48GB	1152GB	20TB	40Gbps	6000Mbps	$.72/hr	($480/m)
Linode	20	64GB	1536GB	20TB	40Gbps	8000Mbps	$.96/hr	($640/m)
Linode	20	96GB	1920GB	20TB	40Gbps	10000Mbps	$1.44/hr	($960/m)

DigitalOcean Pricing Plans

Web Host	CPU Cores	Memory	SSD Disk	Bandwidth	Price
DigitalOcean	1	512MB	20GB	1TB	$5/m
DigitalOcean	1	1GB	30GB	2TB	$10/m
DigitalOcean	2	2GB	40GB	3TB	$20/m
DigitalOcean	2	4GB	60GB	4TB	$40/m
DigitalOcean	4	8GB	80GB	5TB	$80/m
DigitalOcean	8	16GB	160GB	6TB	$160/m
DigitalOcean	12	32GB	320GB	7TB	$320/m
DigitalOcean	16	48GB	480GB	8TB	$480/m
DigitalOcean	20	64GB	640GB	9TB	$640/m

Vultr Pricing Plans as at May 6th, 2015 plans/price change

Web Host	CPU	Memory	SSD Storage	Transfer	Monthly
Vultr	1	768MB	15GB	1.00TB	$5.00
Vultr	1	1024MB	20GB	2.00TB	$7.00 $10.00
Vultr	2	2048MB	40GB 45GB	3.00TB	$15.00 $20.00
Vultr	2	4096MB	65GB 90GB	4.00TB	$35.00 $40.00
Vultr	4	8192MB	120GB 150GB	5.00TB	$70.00 $80.00
Vultr	4	16384MB	250GB	8.00TB	$125.00 Discontinued
Vultr	8	16384MB	250GB 300GB	6.00TB	$149.95 $160.00
Vultr	12	32768MB	400GB	7.00TB	$299.95 Discontinued
Vultr	16	49152MB 32768MB	600GB	8.00TB 10.00TB	$429.95 $320.00
Vultr	24	65536MB	800GB 700GB	9.00TB 15.00TB	$599.95 $640.00
Vultr	1	512MB	160GB 125GB SATA	1.00TB	$5.00
Vultr	1	1024MB	320GB 250GB SATA	2.00TB	$8.00 $10.00
Vultr	1	2048MB	640GB 500GB SATA	3.00TB	$15.00 $20.00
Vultr	2	3072MB	750GB SATA	4.00TB	$30.00
Vultr	2	4096MB	1000GB SATA	5.00TB	$40.00

 

Yeah I am only choosing Vultr for some select locations like Sydney, Tokyo and London usage. Otherwise, it might be better looking for alternatives

 

for $120/m high ram instance

• 32 GB RAM
• 2 CPU Cores
• 40 GB SSD Storage
• 6 TB Transfer
• 40 Gbps Network In
• 1500 Mbps Network Out

Nice and fast there

cpu used ?

might want to add both timed installs to https://community.centminmod.com/threads/post-your-centmin-mod-123-09beta01-install-time-stats.8866/

 

oh thought you mean Linode $120/month instance. So Vultr $120/m instance with dedicated compute Vultr: High Performance SSD Cloud., those use the E3-12xx 3.4Ghz cpus which Vultr use to advertise, they moved to Vultr Dedicated Compute VPS for more dedicated resources